Photogrammetry is a process that uses multiple pictures of a real world object or location to generate a 3D model of that object or location. While this technique has existed almost as long as photography, today software is used to process photographs and generate a digital 3D model with almost no labor from the user.
Photogrammetry requires at least 20-30 photos of an object taken from all possible angles. Any areas of the object that are not photographed can not be modeled. Therefore, it is important to try and photographs all sides, the top, and the bottom of an object.
A 3D model is a digital object that has three-dimensional qualities. 3D Models are created by vertices or points, which are joined to create faces. Together, vertices and faces create a mesh that form the shape of the object being 3D modeled.
3D models can be created in many ways, including:
Photogrammetry - using photographs
Scanning - using laser or structured light
From Scratch - using 3D modeling software to shape and create an object from scratch
Transparent or shiny object don’t work great - Photogrammetry software has a difficult time determining distance and angles in photos of transparent or shiny objects.
Select an object you can photograph from all sides - If you can’t photograph the back, top, or bottom of an object, the software will not be able to model that area. This means your model could have a hole or flat surface in those unphotographed areas.
Use even, diffuse lighting - Having even lighting on your object will allow all areas of the object to photographs well. Avoiding shadows and bright spots in your photos will make it easier for the photogrammetry software to model your object.
Use a background that is all the same color - Having a solid colored background minimizes the need to edit photos and helps prevent the photogrammetry software from pulling background features into your model.